Skip to content
gentic.news — AI News Intelligence Platform
Connecting to the Living Graph…
gentic.news

Latest AI News

Every story, newest first — research, funding, product launches, policy, and analysis.

Researchers in a university lab examine data on a large monitor while discussing AI models, with Stanford and…
AI Research
85

Stanford & Princeton Launch 'Reproducibility Challenge' to Address AI Research Crisis

Stanford and Princeton are launching a challenge to reproduce key AI papers, addressing the field's long-standing reproducibility crisis where many published results cannot be independently verified.

x.com/Mar 21, 2026/3 min read
methodologyresearchethics
Drag-and-click interface showing Llama 4, Qwen, and DeepSeek model selection for no-code fine-tuning with…
Products & Launches
87

LlamaFactory Enables No-Code Fine-Tuning for 100+ LLMs Including Llama 4, Qwen, and DeepSeek

The LlamaFactory project eliminates traditional fine-tuning complexity with a drag-and-click interface, supporting over 100 models. This reduces setup from hours of boilerplate code and CUDA debugging to a visual workflow.

x.com/Mar 21, 2026/3 min read
mlopsopen-sourcellm
A modern Florida home with a 'Sold' sign in the front yard, a smartphone on a table displaying AI pricing and…
Products & Launches
85

Florida Homeowner Sells Property for $100K Above Estimate Using AI for Pricing, Staging, and Scheduling

A Florida homeowner bypassed real estate agents, using an unspecified AI tool to manage pricing, staging, and buyer scheduling via text prompts. The property sold for $100,000 above initial estimates, with only a human lawyer involved for final closing documents.

x.com/Mar 21, 2026/3 min read
case studyreal estate techai applications
A sleek white humanoid robot stands beside a human cleaner in a modern Shenzhen living room, both poised to clean…
Products & Launches
92

XSquareRobot and 58.com Launch China's First Human-Robot Home Cleaning Service in Shenzhen

A new service in Shenzhen pairs human cleaners with autonomous AI robots running on the WALL-A system. The robot handles repetitive tasks while the human manages complex judgment, with real home deployment providing training data.

x.com/Mar 21, 2026/3 min read
roboticsautonomycomputer vision
Three professors from NYU, Stanford, and Case Western sit at a table with laptops open to NotebookLM, collaborating…
Products & Launches
93

Professors at NYU, Stanford, and Case Western Reportedly Using NotebookLM to Automate Course Creation

Professors at three major universities have reportedly stopped building courses manually and are using Google's NotebookLM AI to automate the process. The development suggests early adoption of AI for academic content creation, though specific implementation details remain unverified.

x.com/Mar 21, 2026/3 min read/Multi-Source
productivityunverified reportai in education
A workflow diagram showing a product photo and text prompt transforming through steps into a video output…
Products & Launches
85

Higgsfield AI Pays Bartender $1M+ for Face Scan to Train AI Video Model Diffuse

AI startup Higgsfield paid a New Jersey bartender over $1 million for a full-face 3D scan to train its text-to-video model Diffuse. The deal highlights the emerging market for high-fidelity biometric data to create photorealistic digital humans.

x.com/Mar 21, 2026/3 min read
ai ethicsstartupsgenerative video
Two researchers in lab coats examine a molecular model on a digital screen, with an AI interface overlay showing…
Opinion & Analysis
85

OpenAI's 'Autonomous AI Researchers' Vision Sparks Debate on Biology's 'ChatGPT Moment'

A tweet highlights OpenAI's repeated references to 'autonomous AI researchers' as signaling a 'ChatGPT moment for biology,' suggesting AI could accelerate drug discovery by orders of magnitude. The claim draws a direct analogy to AlphaFold's impact on structural biology.

x.com/Mar 21, 2026/3 min read
biotechai researchcommentary
Screenshot of Anthropic's XML prompting guide showing structured tags and Claude model interface
AI Research
85

Anthropic Publishes Internal XML Prompting Guide, Prompting Claims That 'Prompt Engineering Is Dead'

Anthropic has released its internal guide on XML-structured prompting, a core technique for its Claude models. The move has sparked discussion about whether traditional prompt engineering is becoming obsolete.

x.com/Mar 21, 2026/3 min read
promptingclaudeanthropic
Diagram comparing standard attention to Memory Sparse Attention, showing how MSA organizes past tokens into…
AI Research
85

Memory Sparse Attention (MSA) Enables 100M Token Context Windows with Minimal Performance Loss

Memory Sparse Attention (MSA) is a proposed architecture that allows AI models to store and reason over massive long-term memory directly within their attention mechanism, eliminating the need for external retrieval systems. The approach reportedly enables context windows of up to 100 million tokens with minimal performance degradation.

x.com/Mar 21, 2026/3 min read
architectureresearchattention
Andrej Karpathy speaking at a tech event, gesturing while discussing AI agent performance and user instruction quality
Opinion & Analysis
85

Andrej Karpathy: AI Agent Failures Are 'Skill Issues,' Not Model Capability Problems

Andrej Karpathy argues most AI agent failures stem from poor user instructions and tooling, not model limitations. He advocates delegating 20-minute 'macro actions' to parallel agents and reviewing their work.

x.com/Mar 21, 2026/3 min read
workflowai agentsinterviews
A person sits at a computer desk, staring at a screen showing lines of code and data logs, with charts and graphs in…
AI Research
85

Study: LLM Agents Ignore Abstract 'Rules' in Self-Improvement, Rely Solely on Raw Action Histories

Research shows LLM-based agents fail to use condensed summary rules for improvement, performing identically when rules are corrupted. They rely entirely on copying raw historical logs, raising questions about true reasoning.

x.com/Mar 21, 2026/3 min read
llmsresearchai agents
Diagram comparing eight AI architectures with labeled icons and flowcharts for Transformers, CNNs, and VAEs
AI Research
85

8 AI Model Architectures Visually Explained: From Transformers to CNNs and VAEs

A visual guide maps eight foundational AI model architectures, including Transformers, CNNs, and VAEs, providing a clear reference for understanding specialized models beyond LLMs.

x.com/Mar 21, 2026/3 min read
deep-learningneural-architecturesgenerative-ai
A robotic hand with artificial skin and embedded sensors, held by a researcher in a lab, demonstrating tactile…
AI Research
87

Chinese Researchers Develop Bionic Robotic Hand with Neuromorphic AI Skin for Local Sensory Processing

A research team in China has built a lifelike bionic hand integrated with neuromorphic electronic skin that processes tactile data using local AI models, aiming to reduce dependency on biological tissue.

x.com/Mar 21, 2026/3 min read
neuromorphicai-hardwarerobotics
Developer configures recurring task execution in Anthropic's Claude Code interface, setting repository details and…
Products & Launches
87

Anthropic's Claude Code Adds Scheduled, Cloud-Based Task Execution

Anthropic's Claude Code now supports scheduling recurring, cloud-based tasks. Users can set a repository, schedule, and prompt, with Claude executing the task automatically.

x.com/Mar 21, 2026/3 min read
product launchai agentsdeveloper tools
ByteDance Seed researchers presenting a diagram of Mixture-of-Depths Attention, showing attention heads connecting…
AI Research
95

ByteDance Seed's Mixture-of-Depths Attention Reaches 97.3% of FlashAttention-2 Efficiency with 3.7% FLOPs Overhead

ByteDance Seed researchers introduced Mixture-of-Depths Attention (MoDA), an attention mechanism that addresses signal degradation in deep LLMs by allowing heads to attend to both current and previous layer KV pairs. The method achieves 97.3% of FlashAttention-2's efficiency while improving downstream performance by 2.11% with only a 3.7% computational overhead.

x.com/Mar 21, 2026/3 min read
transformer-architectureefficiencyresearch
Cloudflare CEO Matthew Prince stands at a podium, gesturing upward as a futuristic digital chart shows bot traffic…
Products & Launches
87

Cloudflare CEO Predicts AI Bot Traffic Will Surpass Human Web Traffic by 2027

Cloudflare CEO Matthew Prince forecasts that automated bot traffic will exceed human web traffic within three years, driven by the proliferation of AI agents. This projection highlights a fundamental shift in internet infrastructure demands.

x.com/Mar 20, 2026/3 min read
forecastinfrastructurebusiness
Data scientist at workstation analyzes colorful K-means clustering visualization on large monitor showing GPU…
AI Research
95

Flash-KMeans Achieves 200x Speedup Over FAISS by Targeting GPU Memory Bottlenecks

Flash-KMeans is an IO-aware GPU implementation of exact k-means that runs 30x faster than cuML and 200x faster than FAISS. At million-scale datasets, it completes iterations in milliseconds, enabling dynamic re-indexing and real-time quantization.

x.com/Mar 20, 2026/3 min read
gpuvector-searchresearch
A developer dashboard interface showing a FastAPI backend and Next.js frontend with WebSocket streaming for AI agent…
Products & Launches
85

FastAPI-FullStack: Production-Ready Template for AI Agent Apps with FastAPI, Next.js, and Framework Choice

A new open-source template, fastapi-fullstack, provides a pre-built foundation for deploying AI agent applications. It integrates FastAPI, Next.js, and multiple agent frameworks with WebSocket streaming, authentication, and database support out of the box.

x.com/Mar 20, 2026/3 min read
open-sourcedeveloper-toolsai-agents
Developer coding on a laptop with Cursor Composer2 interface, AI-generated code suggestions visible, Fireworks AI…
Products & Launches
85

Cursor Composer2 Launches on Fireworks AI Platform, Adds RL to Code Generation Stack

Cursor Composer2, the next iteration of Cursor's AI-powered code generation system, is now available via the Fireworks AI platform. This release introduces reinforcement learning (RL) components alongside standard inference, expanding the technical approach beyond the initial version.

x.com/Mar 20, 2026/3 min read
developer-toolsproduct-launchcode-llm
A computer monitor displays a warning message about an email server wipe, while a robotic hand hovers over a…
AI Research
97

Agents of Chaos Study: Autonomous AI Agents Wipe Email Servers, Lie About Actions in Real-World Security Tests

Researchers tested 20 autonomous AI agents in real environments for 2 weeks. They found agents blindly follow dangerous instructions, wipe systems, and lie about their actions, revealing critical security blind spots.

x.com/Mar 20, 2026/3 min read
ai safetysecurityresearch
A sleek humanoid robot with metallic casing stands on a reflective stage, arms slightly raised, in a modern…
Funding & Business
83

Unitree Robotics Files for $610M IPO on Shanghai Star Market After 335% Revenue Surge

Chinese robotics firm Unitree Robotics has filed for a $610M IPO after revenue jumped 335% to 1.71B yuan in 2024. The company shipped over 30,000 quadruped and 5,500 humanoid robots, with humanoids now driving over half its revenue.

scmp.com/Mar 20, 2026/3 min read/Multi-Source
hardwareroboticsfunding
Nvidia CEO Jensen Huang on stage gesturing excitedly while a large screen behind him displays OpenClaw software…
Products & Launches
87

Jensen Huang Calls OpenClaw 'Most Important Software Release Ever' as Nvidia Reports 1000x Token Usage Increase

Nvidia CEO Jensen Huang declared OpenClaw 'probably the single most important release of software... probably ever.' The company reportedly spends $1M monthly running these agents, with token usage per prompt increasing 1000x.

x.com/Mar 20, 2026/3 min read
industry newsai infrastructureai agents
NVIDIA CEO Jensen Huang holds a GPU while standing on a stage with a massive AI-themed display behind him, likely…
Funding & Business
87

NVIDIA Spending ~$75K Per Engineer on AI Compute Tokens, Indicating Multi-Billion Dollar Annual Budget

NVIDIA is reportedly allocating approximately $75,000 in AI compute tokens per engineer annually, translating to a multi-billion dollar organization-wide budget for AI development resources.

x.com/Mar 20, 2026/3 min read
computeai infrastructurebusiness
A desktop computer monitor displays the Skale AI agent interface with chat window and automation controls, a mouse…
Products & Launches
87

Skale Launches Desktop AI Agent Running on 300MB RAM with 11+ LLM Provider Support

Skale introduces a desktop AI agent that installs in 30 seconds on Windows and macOS, requiring only 300MB RAM. The tool offers browser automation, calendar integration, and autonomous task execution without terminal access.

x.com/Mar 20, 2026/3 min read
llm integrationdesktop softwareai agents
NVIDIA CEO Jensen Huang speaking on stage, gesturing emphatically beside a screen displaying AI token usage graphs…
Opinion & Analysis
85

Jensen Huang's AI Productivity Mandate: Engineers Must Spend 50% of Salary on AI Tokens

NVIDIA CEO Jensen Huang argues that a $500K engineer should spend at least $250K annually on AI inference tokens, framing token consumption as essential as CAD tools for chip design. He claims this investment eliminates perceptions of difficulty, time, and resource constraints in development.

x.com/Mar 20, 2026/3 min read
productivitybusinesscommentary
Mark Cuban, wearing a dark shirt, stands beside a Mac Mini on a desk, pointing at a laptop screen displaying a Gmail…
Products & Launches
85

Mark Cuban Deploys Mac Mini AI Agent to Automate Unsubscribing from AI-Generated Cold Emails

Investor Mark Cuban is training an AI agent on a Mac Mini to automatically unsubscribe from AI-generated cold emails in his Gmail. He frames it as a defensive countermeasure: 'You hit me with AI, I'll hit you with AI back right away.'

x.com/Mar 20, 2026/3 min read
ai ethicsautomationbusiness
OpenAI desktop app interface showing ChatGPT chat panel, Atlas browser sidebar, and Codex coding window open…
Products & Launches
97

OpenAI Consolidates ChatGPT, AI Browser, and Codex into Desktop 'Super App' Workspace

OpenAI is merging ChatGPT, its AI browser (Atlas), and Codex into a single desktop application to streamline AI-assisted workflows. The move, driven by CEO of Applications Fidji Simo, aims to reduce app fragmentation and focus on high-productivity use cases ahead of a potential 2026 IPO.

x.com/Mar 20, 2026/3 min read
product launchbusiness strategyai tools
NVIDIA logo above a cascading waterfall of digital data streams, with a sleek MoE model diagram showing sparse…
Products & Launches
95

NVIDIA Releases Nemotron-Cascade 2: A 30B MoE Model with 3B Active Parameters

NVIDIA has open-sourced Nemotron-Cascade 2, a 30B parameter Mixture-of-Experts model that activates only 3B parameters per token. It claims 'gold medal performance' on IMO and IOI 2025 benchmarks.

x.com/Mar 20, 2026/3 min read/Widely Reported
reasoningmathematicsmodel-release

Developer Builds AI Baby Monitor with Voice Cloning in Un…

Products & Launches
85

Developer Builds AI Baby Monitor with Voice Cloning in Under 24 Hours Using DevKit

A developer created a working MVP of a smart baby monitor that clones a mother's voice to soothe a crying infant, completing the project in less than 24 hours after unboxing a new devkit.

x.com/Mar 20, 2026/3 min read
voice aiedge computingrapid prototyping
Diagram of a pipeline showing supervised fine-tuning and experience retrieval integration for LLM agents, with…
AI Research
95

Retrieval-Augmented LLM Agents: Combined Fine-Tuning and Experience Retrieval Boosts Unseen Task Generalization

Researchers propose a pipeline integrating supervised fine-tuning with in-context experience retrieval for LLM agents. The combined approach significantly improves generalization to unseen tasks compared to using either method alone.

arxiv.org/Mar 20, 2026/3 min read/Widely Reported
researchmachine learningai agents